Employing inverse return-to-zero and Manchester formats for uplink wavelength reuse in RoF systems

The inverse return-zero (IRZ) modulation format and Manchester format were investigated as wavelength reusing schemes in the RoF downlink systems. Since both of the modulation formats featured power remaining in the bit slot regardless of the bit value, the downlink optical carrier can be reused as the uplink light source. The performances of two formats were analyzed in the symmetrical and asymmetrical duplex RoF system. It was found that IRZ is more suitable for symmetrical system, and the duty cycle of IRZ pulse is the critical factor on the asymmetrical RoF system performance. However the Manchester format demonstrated the large system tolerance on asymmetrical duplex RoF system with the receiver sensitivity degradation within 1 dB.

Research Highlights
► The other author has no contribution on this work, even grammar revise. Also the reference is revised again. (The reviewer is very much qualified to be the coauthor).
► Paragraph on “IRZ format is well known with large tolerance ….” is rewritten to clarify.
► “In order to generate Manchester pulses…..” The wider IRZ is often used for better dispersion tolerance when RZ is narrow. This paragraph is revised.
► “The baseband signal after MZM is therefore up-converted to the two optical subcarriers……” I only use the first order mixing term ( from the two subcarriers just beside the optical carrier) after photodetection, This paragraph is rewritten to clarify.
► Fig. 3 add the horizontal scale, but I could not generate the vertical scale anymore without software support currently.
► ”An optical interleaver with a channel spacing of 25 GHz” clarify as a commercial applicable interleaver.
► “2046” is a typo, revised as “2048”.
► “BER evaluation ……. , I used Gaussian_ISI in VPI, clarified.
► “In the downlink, as the DC increase, the IRZ pulse width becomes narrower….” The reviewer should resort to the definition of DC on the second page. The reviewer's comment of “one would expect the IRZ pulse width to increase with increasing duty cycle” is wrong. This sentence is rewritten again.
► It is not about inter-channel effect discussed here. If the phase distortion from the two subcarriers dominates, you would get a worse result. Since I get a better result, it is due to less nonlinearity accumulation domination.
► “However, because of the optical filter bandwidth has to be optimized in the Manchester demodulating…” It is rewritten.